Elasticsearch/OpenSearch MCP Server

@cr7258

リポジトリ

基本情報

カテゴリ

データベース

ライセンス

Apache-2.0

トランスポート

stdio

作者

cr7258

Elasticsearch/OpenSearch MCP Server について

A Model Context Protocol (MCP) server implementation that provides Elasticsearch and OpenSearch interaction.

Elasticsearch/OpenSearch MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "elasticsearch-mcp-server": {
      "command": "uvx",
      "args": [
        "elasticsearch-mcp-server"
      ],
      "env": {
        "ELASTICSEARCH_CLUSTERS": "{\"prod\": {\"hosts\": [\"https://prod-es:9200\"], \"api_key\": \"<PROD_API_KEY>\", \"verify_certs\": true}, \"staging\": {\"hosts\": [\"https://staging-es:9200\"], \"username\": \"elastic\", \"password\": \"<STAGING_PASSWORD>\"}}",
        "DEFAULT_CLUSTER": "prod"
      }
    }
  }
}

Elasticsearch/OpenSearch MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Elasticsearch/OpenSearch MCP Server?

A Model Context Protocol (MCP) server that provides Elasticsearch and OpenSearch interaction. It enables searching documents, analyzing indices, and managing clusters through a set of tools. Designed for developers and operators who need AI‑assisted database access.

How to use Elasticsearch/OpenSearch MCP Server?

Install via uvx (automatically from PyPI) or clone the repo and run with uv. Configure environment variables for host, authentication (username/password or API key), and optional multi-cluster settings. Add the configuration to your MCP client (e.g., Claude Desktop) using either the stdio or SSE transport.

Key features of Elasticsearch/OpenSearch MCP Server

Tools for index, document, cluster, alias, and analyzer operations
Multi‑cluster configuration with named clusters and a default target
Authentication via username/password or API key (Elasticsearch)
Option to disable high‑risk write operations
Both stdio and SSE transport support
Bearer token authentication for HTTP transports

Use cases of Elasticsearch/OpenSearch MCP Server

Search and retrieve documents from an Elasticsearch or OpenSearch cluster
Create, update, and delete indices and data streams programmatically
Monitor cluster health and statistics
Analyze text tokenization for query debugging
Manage multiple Elasticsearch/OpenSearch clusters from a single MCP server

よくある質問

What authentication methods are supported?

Basic authentication (username/password) for both Elasticsearch and OpenSearch, and API key authentication for Elasticsearch. API key authentication is recommended for Elasticsearch.

How do I configure multiple clusters?

Set the `ELASTICSEARCH_CLUSTERS` or `OPENSEARCH_CLUSTERS` environment variable to a JSON object with named clusters. Alternatively, use `ELASTICSEARCH_CLUSTERS_FILE` or `OPENSEARCH_CLUSTERS_FILE` to point to a JSON file. Every tool accepts an optional `cluster` parameter.

Can I disable all write operations?

Yes, set `DISABLE_HIGH_RISK_OPERATIONS=true` to hide all write tools (create/delete index, document operations, alias changes, etc.). You can also disable specific operations with `DISABLE_OPERATIONS`.

Is SSL verification enabled by default?

No, `VERIFY_CERTS` defaults to `false`. You can set it to `true` to verify SSL certificates.

How do I secure the server over HTTP transports?

Set the `MCP_API_KEY` environment variable to enable Bearer token authentication for SSE and Streamable HTTP transports. Without it, the server is accessible without authentication. The `stdio` transport does not require this.